Skip to Main Content
The performance of MPEG video transmission over error-prone channels is limited by the channel noise. An efficient error concealment (EC) scheme is essential for diminishing the impact of transmission errors in a compressed video, A number of EC techniques have been developed to combat the transmission errors. However, the previous techniques are always inefficient when the motions of video object are fast or complex. This paper proposes a novel adaptive EC algorithm to conceal the error for block and motion-compensation based video coding systems. The proposed EC method employs an unsupervised artificial neural network (ANN) model, i.e. self-organizing map (SOM), as a predictor to estimate the motion vectors of the damaged macroblocks (MBs). Then the estimated motion vectors were utilized to reconstruct the damaged MB by exploiting the spatial information from reference frames based on the boundary matching criterion. Because of the SOM has a great capacity for visualizing and interpreting high-dimensional data sets, the estimation model proposed herein can exploit the nonlinearity property of the SOM to estimate lost motion vectors more accurately. Computer simulations show that the visual quality and the PSNR evaluation of reconstructed frames are significantly improved by using the proposed EC algorithm. Thus, the proposed algorithm is expected to be practical for motion vector compressed video in error-prone networks.